Management and Future Technologies | Volume 2, Issue 2 | № 006

Management and Future Technologies № AI methods journal.umft.uz June 2025 Application Area 44 Cardiovascular disease prediction 45 Diagnosis and prediction of active tuberculosis 46 47 Logistic Regression (LR) Multivariate Logistic Regression (MLR) Preoperative diagnosis / combined models 83–93% XGBoost (XGB) General lung cancer risk model (considering SNPs and others) Cancer risk prediction in men Cancer risk prediction in smokers Lung adenocarcinoma (ADC) prediction Cancer risk prediction Cancer risk prediction 79.10% 78.50% 69.90% 83% 83% Heart disease prediction (Cleveland dataset) 82–99% Cardiovascular disease prediction 79.12% (DatasetI), 99.03% (DatasetII) Cancer risk prediction 82% 50 52 58 59 85–92% 83.30% 49 53 54 55 56 57 Accuracy (AUCROC) 85.71% (DatasetI), 82.14% (DatasetII) Lymphoma diagnosis (fever of unknown origin) (FUO) [4] Lymphoma diagnosis (fever of unknown origin) (FUO) [7] Cervical cancer classification General lung cancer risk model (considering SNPs and risk factors) Squamous cell lung cancer (SCC) prediction 48 51 Volume 2, Issue 2 Extreme Gradient Boosting (XGB) 60 Gradient Boosting (GB) 61 62 63 64 65 Cervical cancer diagnosis Colorectal cancer diagnosis Intuitionistic Nasopharyngeal cancer diagnosis fuzzy sets (IFS) Breast cancer diagnosis Lung cancer diagnosis 86% 93.60% 98% 74.20% 75.90% 100% 99.89% 99.72% 100% 100% This diagram presents an analysis of the accuracy of machine learning algorithms in medical diagnostics, revealing that the Random Forest (RF) algorithm demonstrated the highest performance, 261

Volume 2, Issue 2 262 June 2025 Management and Future Technologies journal.umft.uz achieving 100% accuracy in cervical cancer classification. Comparable effectiveness was shown by AdaBoost and Decision Tree (DT) algorithms, highlighting the high reliability of ensemble methods. High accuracy rates were also recorded for the Multilayer Perceptron (MLP) — 99.17%, Support Vector Machine (SVM) — 99.5%, and the hybrid Majority Voting approach — 99.23%. Algorithms such as XGBoost, CatBoost, and LightGBM demonstrated stable efficiency in predicting oncological and cardiovascular diseases.Thus, Random Forest (RF) can be considered the most effective method for early diagnosis (Figure 7). Figure 7. Chart of algorithms with the highest disease prediction accuracy. The analysis of the diagram, compiled based on the reviewed sources, demonstrates the high potential of machine learning algorithms in cancer diagnosis and prognosis tasks. The highest accuracy (100%) was achieved by the Random Forest, AdaBoost, and Decision Tree algorithms in the classification of cervical cancer, breast cancer, and lymphomas, confirming their reliability in clinical practice. High performance was also shown by the Support Vector Machine (SVM), Convolutional Neural Network (CNN), and Multilayer Perceptron (MLP) algorithms, with accuracy ranging from 99.17% to 99.51%.The GPT-like CHIEF model achieved 94% accuracy in multi-cancer diagnosis (covering 19 types of cancer), highlighting the potential of transformer-based architectures. The lowest, yet still substantial, results were demonstrated by LightGBM (83.6%) and CatBoost (85.2%) in cancer risk prediction tasks among patients with diabetes (figure 8).

Management and Future Technologies journal.umft.uz June 2025 Volume 2, Issue 2 263 Figure 8. Accuracy of AI models in cancer diagnosis. The pie chart presents a comparative assessment of the accuracy (or AUC) of machine learning models for the detection of various diseases. The highest value (100%) was obtained for cervical cancer classification using UCI data, indicating a perfect match between the model’s predictions and actual diagnoses. A similarly high result — 99.8% — was recorded for the identification of lymphoma in patients with fever of unknown origin (FUO), particularly on the training dataset. Other pathologies also demonstrated similarly high accuracy levels: 98.05% in cardiovascular disease prediction and 97% when working with EHR data for cervical cancer one day before diagnosis.The lowest, yet still high, result — 91% — was recorded in diabetes prediction (figure 9). Figure 9. Classification accuracy Conclusion

Volume 2, Issue 2 June 2025 Management and Future Technologies journal.umft.uz Conclusion 264 The analysis of scientific researches on the application of machine learning methods in medical diagnostics demonstrates the high effectiveness of modern algorithms, particularly in the early detection of oncological diseases. In this study, a dataset of 1000 cervical cancer cases was collected to support the research. Among the analyzed methods, the Random Forest (RF) algorithm exhibited the most outstanding results, achieving 100% accuracy in diagnosing cervical cancer and demonstrating high stability in predicting the survival rates of oncology patients. Ensemble methods such as XGBoost and AdaBoost, along with the Support Vector Machine (SVM) and Decision Tree (DT), also Intuitionistic fuzzy set algorithms, confirmed their reliability and interpretability for clinical use. Special attention should be given to the work of Chan S.W. and Abdullah L. [16], who proposed the use of Intuitionistic Fuzzy Sets for cancer diagnosis. The application of Euclidean and Hamming distances enabled high stability and reproducibility in classification results, opening new prospects for integrating fuzzy logic into clinical diagnostic systems. Thus, the integration of algorithms such as Random Forest, SVM, and models based on IFS represents a promising direction for the development of reliable intelligent decision support systems in medicine.

Management and Future Technologies journal.umft.uz June 2025 Volume 2, Issue 2 References 1. Nafiseh Ghaffar Nia, Erkan Kaplanoglu, Ahad Nasab. Evaluation of artificial intelligence techniques in disease diagnosis and prediction. Springer, 30 January 2023. https://link.springer.com/ article/10.1007/s44163-023-00049-5 2. Rahman M.M., Asyhari A.T., Ahmed F. Influence of Optimal Hyperparameters on the Performance of Machine Learning Algorithms for Predicting Heart Disease. MDPI Processes, 1 March 2023. https:// www.mdpi.com/2227-9717/11/3/734 3. Muluken Belay, Worku Necho, Misganaw Fikadu. Application of supervised machine learning algorithms for classification and prediction of type-2 diabetes disease status in Afar regional state, Northeastern Ethiopia 2021. Nature Scientific Reports, 13 May 2023. https://www.nature.com/ articles/s41598-023-34906-1 4. Zhou L., et al. Machine learning‑based prediction of gastroparesis risk. Springer, 27 September 2024. https://link.springer.com/article/10.1007/s12672-024-01355-9 5. Lim J.H., et al. Prediction of fellow eye neovascularization in type 3 macular neovascularization (Retinal angiomatous proliferation) using deep learning. PLOS ONE, 30 October 2024. https:// journals.plos.org/plosone/article?id=10.1371/journal.pone.0310097 6. Qiu L., et al. Plasma immune profiling combined with machine learning contributes to diagnosis and prognosis of active pulmonary tuberculosis. Journal of Tuberculosis, 04 July 2024. https://www. tandfonline.com/doi/full/10.1080/22221751.2024.2370399 7. Ma H., et al. Distinguishing lymphoma from benign lymph node diseases in fever of unknown origin using PETCT radiomics. EJNMMI Research, 13 November 2024. https://ejnmmires.springeropen. com/articles/10.1186/s13550-024-01171-w#Sec12 8. Fekri S., et al. Using machine learning for predicting cervical cancer from Swedish electronic health records by mining hierarchical representations. PLOS ONE, 21 August 2020. https://journals.plos. org/plosone/article?id=10.1371/journal.pone.0237911 9. Harvard Medical School. 96% Accuracy: Harvard Scientists Unveil Revolutionary ChatGPT-Like AI for Cancer Diagnosis. SciTechDaily, 17 October 2024. https://scitechdaily.com/96-accuracyharvard-scientists-unveil-revolutionary-chatgpt-like-ai-for-cancer-diagnosis/ 10. Barz M., et al. Application of Machine Learning Methods to Improve the Performance of Ultrasound in Head and Neck Oncology: A Literature Review. MDPI Cancers, 28 January 2022. https://www. mdpi.com/2072-6694/14/3/665 11. Lakshmi V., Radha M., Babu C.S. A Model for Predicting Cervical Cancer Using Machine Learning Algorithms. MDPI Sensors, 29 May 2022. https://doi.org/10.3390/s22114132 12. Li J., et al. Prediction of lung cancer risk in Chinese population with genetic-environment factor using extreme gradient boosting. Cancer Medicine, 24 April 2022. https://onlinelibrary.wiley.com/ doi/10.1002/cam4.4800 13. Saravanan K., et al. Prediction and classification of skin melanoma cancer using active hybrid machine learning technique. Journal of Physics: Conference Series, 25 February 2023. https://doi. org/10.1088/1742-6596/2580/1/012039 14. Qiu X., et al. A machine learning prediction model for cancer risk in patients with type 2 diabetes based on clinical tests. Technology and Health Care, 1 May 2024. https://journals.sagepub.com/doi/ full/10.3233/THC-230385 15. Wu H., et al. Development and validation of a deep learning model for predicting postoperative survival of patients with gastric cancer. BMC Public Health, 6 March 2024. https://bmcpublichealth. biomedcentral.com/articles/10.1186/s12889-024-18221-6 265

Volume 2, Issue 2 266 June 2025 Management and Future Technologies journal.umft.uz 16. Chan S. W., Abdullah L. Distance Measures of Intuitionistic Fuzzy Sets and its Application to Cancer Diagnosis. ASM Science Journal, 2020. https://doi.org/10.32802/asmscj.2020.sm26(5.1) 17. Khidirova Ch., Sadikova Sh., Jabborova N., Sadikova F. Features of Intuitionistic Fuzzy Logic Application in Software Algorithms. // Lecture Notes in Networks and Systems (LNNS), vol.718 (WCIS-2022). Vol. 1. Springer - 2024. -P. 88-95. https://doi.org/10.1007/978-3-031-51521-7_13 18. Khidirova Ch., Jabborova N. Importance of fuzzy logic methods in solving problems of medical diagnosis and prognosis. // International scientific journal science and innovation. Special issue: “Modern problems and prospects of development of energy supply of digital technology facilities”. 2024. -P.224-229. https://doi.org/10.5281/zenodo.10721297 19. Khidirova Ch., Ruzibaev O., Shoimova S. Intelligent System of Diagnosing and Predicting Cardiovascular Diseases. // International Conference on «Information Science and Communications Technologies». 2022. - 4 p. https://doi.org/10.1109/ICISCT55600.2022.1014680

Management and Future Technologies Authors Makhsudov Mokhirbek Andijan State Technical Institute, Andijan, Uzbekistan; mohirbek2702@mail.ru, Xalimjanov Abduqunduz Andijan State Technical Institute, Andijan, Uzbekistan; xalimjonov1234@gmail.com Correspondence: xalimjonov1234@ gmail.com Copyright: © 2025 by the authors. Ushbu maqola Creative Commons Attribution (CC BY) litsenziyasi shartlari asosida tarqatiladigan ochiq foydalanish maqolasi hisoblanadi (https://creativecommons.org/ licenses/by/4.0/). journal.umft.uz June 2025 Volume 2, Issue 2 SIMULATION ANALYSIS OF IMPROVING THE ENERGY EFFICIENCY OF SINGLE-PHASE AND THREE-PHASE RECTIFIERS IN DIGITAL METHODS A bstract: This article discusses the structure of onephase and three-phase rectifiers, their principles of operation, voltage, and current. Diagrams, corrective coefficients, advantages and disadvantages, their energy methods to improve the effectiveness of the project, and also methods for grinding output parameters of one- and threephase rectifiers, time charts and the impact on network quality indicators were considered. Key words: bridge scheme, capacity, current, filter, energy efficiency, rectifier, single-phase, three-phase, voltage. 267

Volume 2, Issue 2 June 2025 Management and Future Technologies journal.umft.uz Introduction 268 Two Semi-Closed Rectifier Circuits – Advantage: Rectifier in this Circuit Compared to a single half-circuit rectifier, its pulse is half as small. One capacitance of the capacitor at the same ripple coefficient as that of the half-loop rectifier can be twice as much. Disadvantages: complexity of the design of the transformer and in the transformer Jump to search Bridge chain straightener – an advantage: compared to a semi-circular straightener in the bridge scheme, the level of pulmonation is 2 times less, high efficiency of work, the transformer is used rationally, its design power is reduced [1]. When the level of pulsation is the same in relation to the two half-cycle schemes, a simpler transformer design is used in the bridge scheme. The reverse voltage of the valves may be considerably lower than in previous schemes. Disadvantage: the increase in the number of valves and the necessity to shunt each valve to adjust the reverse voltage. The voltage doubling scheme – Advantages: the secondary winding of the transformer can be calculated for a practically lower voltage. Figure 1. Simulation circuit of Single phase rectifier with filter element. Disadvantages: somewhat more currents pass through the rectifier valves[2,3]. The level of pulsation is higher than in the two half-cycle schemes. To improve the power factor of three-phase rectifiers, a null valve scheme (similar to two half-cycle schemes) is employed. In a three-phase scheme, the current flowing through the transformer windings has an alternating character and possesses equal values in both positive and negative half-cycles. This situation prevents the magnetization of the transformer cores and is considered one of the main characteristics of this scheme [4-6]. This indicates that the average current is present in both the bridge scheme of Id and the midpoint scheme. The highest limit of the same values is permissible, and the cooling conditions of the thread are applicable. The same valves can be selected [7, 8-10]. Values of the Uk coefficient by voltage demonstrate that at the same values of Oud in a bridged scheme, it is superior to that of the midpoint, as the voltage to the valve is higher at the midpoint. In low voltage circuits, the midpoint is

Management and Future Technologies journal.umft.uz June 2025 Volume 2, Issue 2 more advantageous, since the total voltage drop across the valves is half as much. Multiphase correction circuits are supplied from a three-phase alternating current network and are primarily utilised in medium and high-power devices. Multiphase schemes enable the resolution of the following issues: ‒ reduction of the correct voltage pulses; ‒ improvement of the harmonic composition of the current supplied from the mains; ‒ decrease in the sample (calculated) power of the transformer. In high-power rectifiers, a simple inductive filter is typically employed, as the frequency of output voltage pulses is not excessively large. In inductances, the pulsation coefficient in the load chain is also determined by coefficients [11]. It turns out that only a chain that is linear with a constant organizer of the current will be suitable for carrying out a straightening process. In order for the corrector effect beyond it to be sufficient, this invariant organizing current must be large. Such a property has a non-linear chain in which an element with a Valve property in the composition of lamp and semiconductor circles such as a gasotron, a tratron has this property. Therefore, straighteners are assembled from these elements. Since the vibration (input voltage) acting on the chain in the rectifier has a sufficiently large amplitude, the rectifier element graph consists of a straight line. The rectification scheme will have two different single and two semi-periods. The alternating current is fed through the transformer to the D diode. Since the diode is a oneway conductor, the current from R is only uttered in the positive half-period. The current and output voltage in the chain will be continuous. Such a current voltage is called a pulsating current or voltage. A constant part of the corrected voltage. For a single semicircular rectifier U 0 0.45U 2 (1) where U2 is the value of the AC voltage given to the diode; Uo is the rectified voltage. The inverse voltage is in the negative half-VR to the diode lightning voltage. For a single semicircular rectifier U R 3.14U 0 (2) where UR is the maximum value of the reverse voltage. Experimental research When analyzing multi-phase circuits, it is as if a single phase. As in the analysis of circuits, we consider the valves and transformer ideal (without waste). The principle of Simulation circuit of Single phase rectifier with filter element of the straightener is shown in Figure 1. Below are the basic calculation ratios of a rectifier with a three-phase zero output wire quoted: 269

Volume 2, Issue 2 270 June 2025 Management and Future Technologies journal.umft.uz Table 1. Calculation ratios of a rectifier Ud=1.17 E2 Uakmax=2.09Ud E2=0.855Ud I2=Id√3 Ia=Id√3 Iamax=Id II=Id√2 / (3KT) PT=1.35Pd The primary turbulences of the transformer are both "star" and "triangle" and can also be connected in the form of. Secondary swells are only in the form of a "star". Jump to search. Jump to navigation. Jump to the free ends of the secondary loop. The attached anodes or cathodes serve as the other pole of the output voltage. Jump to navigation. The direction of the valve connection determines the output voltage polarity. The pulsation coefficient of the voltage is q=2/(m2 ‒ 1). In this case, the three-phase when the rectifier has m=3, q = 0.25. In this case, the already existing boot chain. Even in small inductances, the loading current is very well polished [12]. Six-phase correction of transformer circuit with midpoint output wire. Disambiguation pages Each rod of a three-phase transformer. There is one primary and two secondary windings. It can be connected in both a star configuration and a delta configuration. The plumes form a six-phase star. The six-phase voltage formed by the secondary circuits of the transformer. The vector diagram of the system is presented in Figure 2. Voltage ratio of each phase. Vector 60 from previous [13]. The electric turned a corner. The zero point of the star forms a negative pole of the loading chain. Cathodes have a common point. The positive potential is higher than that of others. Figure 2 shows the correction scheme of a three-phase bridge. This scheme is also referred to as Larionov's scheme. The primary and secondary circles of the transformer are triangular or asterisk shaped and can also be connected in that form. In principle, the circuit should function even without a transformer. It is possible. ‒ Cathode group, in which cathodes (VD1, VD3, VD5) are combined; ‒ anode group, in which anodes (VD2, VD4, VD6) are combined. The common points of the valves are connected to the supply source (with the secondary circuit of the transformer as shown in the diagram). At any given moment, the anode potential of the valve conducting the current is at its maximum (in the cathode group). Therefore, The instant voltage at the rectifier output depends on the voltage of the phase line that's connected to the load through the open valves at that time. The aforementioned indicators determine the use of the scheme in medium and high-power devices. Thus, The energy efficiency goes up from a single-phase one-and-a-half-circuit to a three-phase rectifier circuit in a similar way.[14]. Active capacity filter (R-C) and inductive capacity filter (L-C) small loading vines. That is, as the loading current increases, the voltage pulsation increases. The advantage of the R-C filter is that the voltage level does not increase in transient processes, nor does it decrease. The disadvantage of

Management and Future Technologies journal.umft.uz June 2025 Volume 2, Issue 2 the filter is that it affects the straightener, which is why the introduction of security elements is necessary. The inductive-capacitive (L-C) grinding filter is used in large loading currents. The advantage is small dimensions, the grinding coefficient depends on the load current changes. Not related. The disadvantage is that the voltage load during transient processes is large, and the time to stagnate from the transient process is enormous. Research results Resonant grinding filters. The corrected voltage of the rectifiers is close to the first harmonic in terms of the degree of the variable component used when it belongs. They are also sensitive to high voltage harmonics and are used for non-disambiguation devices. The feeding voltage frequency is significant. The use of such filters in the event of a large change in voltage frequency is impossible. A change in the loading current causes a change in the circuit inductance, which will also lead to a decrease in the grinding coefficient. To maintain the uniform continuity of inductance, a throttle or a reverse linking loop is introduced. This is due to the increase in the size of the filter, and useful work leads to a decrease in the capacity of the work, which is why the loading current of such filters remains constant. It is recommended only when it pertains to different harmonics of voltage from the first. Additional reactive elements are used for removal. Compared to the grinding filters, these filters are neither large nor coarse, nor do they have a large useful work capacity. A grinding active filter. Active filters are common in small-power devices with a number of advantages [15,16]. These benefits include: High quality and energy indicators; Wide range of frequency; Simplicity of design; The dependence of the coefficient of grinding on the loading current is minimal. Magnetic fields are small due to a lack of inductance in the filter scheme; There is an absence of dangerous modes in the event of transient processes, i.e. boot, so that when it is "turned off", the voltage load is not generated. Disadvantages include: The increase in loading current due to increased waste in the transistor will reduce the efficiency of the device; There is a need to protect the transistor during transient processes. The principle of operation of the active filter is based on the properties of the transistor to resist different currents to alternating and alternating currents. There are two methods for creating filters. The first method is that the transistor is connected according to the common collector scheme [17]. In the general collector scheme, when the value of the base current is constant, the collector current's dependence on the voltage value of the collector-emitter pass is minimal, and this smoothing 271

Volume 2, Issue 2 272 June 2025 Management and Future Technologies journal.umft.uz of pulsations in the wire is achieved by an RC filter on the base circuit. In the second method, the transistor is connected according to a common base scheme (Figure 1). The operating mode of the transistor is determined by the value Rb, while the smoothing effect of the circuit R1 and C1 is defined by it. This circuit stabilises the emitter current if R1C1 is >> Tn (Tn‒ pulsation cycle). In this mode, the transistor has a large differential and a small static resistor. The smoothing coefficient S is calculated in the same manner as in the RC passive filter. S R 1 C f p R R H 1 R H (1) Single phase - and three-phase rectifiers consume a non-sinusoidal current from the network. The harmonic composition of the current depends on the parameters of the correction scheme and the grinding filter. For example, the harmonic composition of the consumed current is one correction scheme; if it is of different types ‒ inductive or capacitive filters, it also differs. If the power of the network generator is equal to the capacity of the fully loaded rectifier or more than that, then the rectifier is not the same as the rectifier. The effect of non sinusoidally on the voltage of the feeder network is practically unknown [18]. The higher the values of generator and rectifier capacities, the higher the current that causes a breakdown (scaffolding) in the voltage of the supply network of harmonics intensifies. Voltage Pattern Distortion is usually a voltage disorder. It is valued by the coefficient of harmonica. k U 2 n 2 U nm U 1m (2) Where in U1m is acceleration of the first harmonic voltage amplitude and Unm is the amplitude of the high harmonic amplitude. The value of this kτU coefficient is determined by the established standards for network voltage and generally does not exceed 5%.

Management and Future Technologies journal.umft.uz June 2025 Volume 2, Issue 2 273 Figure 2. Sinusoidal harmonic oscillation graph Form distortion due to high harmonics in mains voltage increases scaffolding in reactors and transformers, and cable combined in the field, which will lead to additional waste of electricity at the generator and communication, automation, computing, etc., which differ in the operation of electronic equipment. This leads to breakdowns. To address this, it is necessary to reduce the high harmonics in the mains voltage, and special measures are required to reduce the value of the Uk coefficient. Conclusions To determine the harmonic coefficient Uk by voltage Equivalent Replacement Scheme of the Corrector and Feeding Network Form of mains voltage from the operation of straighteners The occurrence of these devices is mainly caused by the internal network supplying these devices. Voltage drops in the resistors (resistors of generators and transformers, and others). In an electrical system, a straightener is usually used as a generator of current harmonics [19]. In this case, the rectifier is equivalent as a source of various harmonics of the current can be specified in the scheme. Figure 2 shows the rectifier with the current source InΣ replaced [20,21]. The time diagram in Figure 2 shows that the single-phase rectifier for the idealized scheme (ωL=∞ and the switching angle γ=0) is 1 until us. From voltage, it lags behind α angles with a value of φ1. Device power coefficient χ = v cos α. The angles φ1 and α of a three-phase bridge circuit rectifier are both equal, and the power coefficient is determined as above. Coefficients of distortion for straight-angle consumer current: v� �√� ‒ for single-phase bridge scheme. �

Volume 2, Issue 2 274 June 2025 Management and Future Technologies journal.umft.uz ଷ v ൌ ‒ for a three-phase bridge circuit. With the increase in the control angle α, the rectifier's grid consumption power of the reagent Q increases, and its coefficient of power decreases, i.e. it will get worse. This phenomenon is caused by a controlled rectifier reducing the technical and economic characteristics. Therefore, in practice, measures to increase the power coefficient of controlled straighteners will be observed.

Management and Future Technologies journal.umft.uz June 2025 Volume 2, Issue 2 References 1. Bibave, R., Thokal, P., Hajare, R., Deulkar, A., William, P., & Chandan, A. T. (2022, March). A comparative analysis of single phase to three phase power converter for input current THD reduction. In 2022 International conference on electronics and renewable systems (ICEARS) (pp. 325-330). IEEE. 2. Gonçalves, J. T., Valtchev, S., Melicio, R., Gonçalves, A., & Blaabjerg, F. (2021). Hybrid three-phase rectifiers with active power factor correction: A systematic review. Electronics, 10(13), 1520. 3. Shneen, S. W., Ahmedhamdi, R. T., & Al-Ghezi, M. K. (2022). A-review of Simulink for single-phase rectifier. Int J Adv Appl Sci ISSN, 2252(8814), 8814. 4. da Cruz Ferreira, R., Dias, S. M. A., Rocha, N., da Silva, E. R. C., & Melo, V. F. B. (2023). Predictive Control for a Single-Phase to Three-Phase Converter with Two-Parallel Single-Phase Rectifiers. Eletrônica de Potência, 28(4), 324-336. 5. Chen, Z. G., Lei, T., Hua, Z. J., Ugli, B. Z. U., & Ugli, A. R. Y. (2024, October). Short-term photovoltaic power generation prediction based on VMD-IGWO-LSTM. In 2024 IEEE 4th International Conference on Digital Twins and Parallel Intelligence (DTPI) (pp. 682-686). IEEE. 6. I. K. Siddikov, M. A. Anarbaev, A. A. Abdumalikov, M. A. Sobirov, M. T. Maxsudov and I. M. Xonturaev, “Technological aspects of modelling and research of smart grid,” 2019 International Conference on Information Science and Communications Technologies (ICISCT), Tashkent, Uzbekistan, 2019, pp. 1-5, doi: 10.1109/ICISCT47635.2019.9011936. 7. Motegi, S. I., & Nishida, Y. (2021, September). A New Three-Phase Diode Rectifier with Improved Characteristics. In 2021 23rd European Conference on Power Electronics and Applications (EPE’21 ECCE Europe) (pp. P-1). IEEE. 8. I.K. Siddikov, M.A. Anarbaev, A.A. Abdumalikov, A. B. Abubakirov, M.T. Maxsudov and I.M. Xonturaev, “Modelling of transducers of nonsymmetrical signals of electrical nets,” 2019 International Conference on Information Science and Communications Technologies (ICISCT), Tashkent, Uzbekistan, 2019, pp. 1-6, doi: 10.1109/ICISCT47635.2019.9011981. 9. Makhsudov, M., & Karimjonov, D. (2023). Research algorithm and dynamic characteristics of threephase electromagnetic current transducers in filter-compensation devices. In E3S Web of Conferences (Vol. 461, p. 01056). EDP Sciences. 10. Mokhirbek Makhsudov, Dilyorbek Karimjonov, Akmaljon Abdumalikov, Akhmadbek Jalilov, Mamurjon Yigitaliyev; Method of determination current and power factor based on the output signal. AIP Conf. Proc. 27 November 2024; 3244 (1): 060009. https://doi.org/10.1063/5.0241566. 11. Dilyorbek Karimjonov, Mokhirbek Makhsudov, Abduqunduz Xalimjanov, Dilshodbek Abdukhalilov, Durbek Axmedov, Xolmirza Mamajonov; Modeling the structures of three-phase asynchronous motor reactive power variations using electromagnetic transducers. AIP Conf. Proc. 27 November 2024; 3244 (1): 060018. https://doi.org/10.1063/5.0241567. 12. I. K. Siddikov, A. A. Abdumalikov and M. T. Makhsudov, “Modeling and Research Multiphases Signal Transducers of Power Control Systems,” 2020 International Conference on Information Science and Communications Technologies (ICISCT), Tashkent, Uzbekistan, 2020, pp. 1-4, doi: 10.1109/ ICISCT50599.2020.9351482. 13. Siddikov, I.Kh., Malikov, A., Makhsudov, M.T., Boikhanov, Z.U., Uzaqov, R. Study of the Static Characteristics of the Secondary Stator Voltage Converter of the Currents of an Induction Motor, AIP Conference Proceedings 2432, 020003 (2022); https://doi.org/10.1063/5.0089681. 14. Muhiddin Atajonov; Development of technology for the development of highly efficient combinations of solar and thermoelectric generators. AIP Conf. Proc. 11 March 2024; 3045 (1): 020011. https:// doi.org/10.1063/5.0197733. 275

Volume 2, Issue 2 276 June 2025 Management and Future Technologies journal.umft.uz 15. Atajonov, M. O., Nimatov, S. J., Rakhmatullaev, A. I., & Sadullaev, A. B. (2023, January). Formalization of the dynamics of the functioning of petrochemical complexes based on the theory of fuzzy sets and fuzzy logic. In AIP Conference Proceedings (Vol. 2552, No. 1). AIP Publishing. https://doi. org/10.1063/5.0112403 16. I. Kh. Siddikov, A. B. Abubakirov, A. U. Djalilov, T. U. Kurbaniyazov, A. A. Abdumalikov; Statistical descriptions of multiphase current sensers of reactive power control systems in renewable power supply power systems. AIP Conf. Proc. 23 June 2023; 2789 (1): 060002. https://doi.org/10.1063/5.0145430 17. I. Kh. Siddikov, Kh.A. Sattarov, A.A. Abdumalikov; The static characteristics of primary current transducers of current of specific electrical loads of renewable power sources.AIP Conf. Proc.15 March 2023; 2612 (1): 050002. https://doi.org/10.1063/5.0115211 18. Siddikov, Ilkhomjon and A. A. Abdumalikov. “Modeling and research signals conversion proceses of multhipase power measure and control devices.” 2020 International Conference on Information Science and Communications Technologies (ICISCT) (2020): 1-4. 19. Atajonov, M., Mamarasulov, Q., Dexqonboyev, O., & Xakimov, T. (2024, November). Research of solar panel model based on photovoltaic module. In AIP Conference Proceedings (Vol. 3244, No. 1). AIP Publishing. https://doi.org/10.1063/5.0241785. 20. Atajonov, M., Mamarasulov, Q., Zaripov, O., Nimatov, S., & Bo‘riyev, U. (2024, November). Study of solar photoelectric plant in MATLAB (simulink) package. In AIP Conference Proceedings (Vol. 3244, No. 1). AIP Publishing. https://doi.org/10.1063/5.0241783 21. Siddikov I.K., Abubakirov A.B., Djalilov A.U., Kurbaniyazov T.U., Abdumalikov A.A., AIP Conference Proceedings, 2789, 060002, (2023)

Management and Future Technologies Mualliflar Axmedov Alimardon Turopova Xilola Akmuradov Baxtiyor “University of Management and Future Technologies” universiteti Kommunikatsiya va raqamli texnologiyalar kafedrasi, b.u.akmuradov@gmail.com Mas’ul: b.u.akmuradov@gmail.com Copyright: © 2025 by the authors. Ushbu maqola Creative Commons Attribution (CC BY) litsenziyasi shartlari asosida tarqatiladigan ochiq foydalanish maqolasi hisoblanadi (https://creativecommons.org/ licenses/by/4.0/). journal.umft.uz June 2025 Volume 2, Issue 2 TELEKOMMUNIKATSIYA 277 TARMOQLARIGA BO‘LADIGAN KIBERHUJUMLARNI ANIQLASH VA OLDINI OLISH TIZIMLARINI TADQIQ QILISH A nnotatsiya: Mazkur tadqiqot ishi telekommunikatsiya tarmoqlarida yuzaga kelishi mumkin bo‘lgan xavfxatarlar va zamonaviy kiber hujumlarga qarshi kurashish tizimlari o‘rganilgan. Tadqiqotda xususan, Snort va Suricata tarmoq hujumlarini aniqlash tizimlarining yuqori tezlikdagi tarmoqlarda ishlash samaradorligi baholangan. Ushbu tizimlar yuqori yuklamali sharoitlarda eksperimental sinovdan o‘tkazilib, ularning aniqlik, tezlik va resurslardan samarali foydalanish ko‘rsatkichlari chuqur tahlil qilingan. Olingan natijalar asosida zamonaviy kiber hujumlarga qarshi samarali strategiyalar ishlab chiqish bo‘yicha tavsiyalar beriladi. Mazkur tadqiqot kiber xavfsizlik sohasidagi ilmiy adabiyotga yangi bilim va tajribalar qo‘shishi bilan birga amaliyotda qo‘llash uchun foydali tavsiyalarni ham beradi. Kalit so‘zlar: kiberhujum, axborot xavfsizligini, telekommunikatsiya tarmoqlari, DoS, DDoS, IDS, IPS, Snort, Suricata.

Volume 2, Issue 2 June 2025 Management and Future Technologies journal.umft.uz Kirish 278 Ushbu tadqiqot ishi telekommunikatsiya tarmoqlarida yuzaga kelishi mumkin bo‘lgan xavfxatarlar va kiber hujumlarni tahlil qilishga qaratilgan bo‘lib, kiber xavfsizlik sohasida dolzarb muammolarni yoritadi. Bugungi kunda kiber hujumlar ko‘payib borayotganligi, yangi texnologiyalar va internet xizmatlarining rivojlanishi bilan bog‘liq ravishda, davlat tashkilotlari uchun tahdidlar yanada kuchaymoqda. Telekommunikatsiya tarmoqlari ham kiber hujumlarga nisbatan zaifligi bilan ajralib turadi, bu esa ularning ishonchliligini va xavfsizligini ta’minlash masalasini dolzarb qilmoqda. Telekommunikatsiya tarmoqlariga bo‘ladigan hujumlarni aniqlash va oldini olish tizimlarini tadqiq qilish mavzusi bo‘yicha qator normativ-huquqiy hujjatlar mavjud bo‘lib, ular ushbu yo‘nalishda ilmiy va amaliy izlanishlar olib borish uchun muhim asos bo‘lib xizmat qiladi. Jumladan, O‘zbekiston Respublikasi Prezidentining 2021-yil 12-fevraldagi PQ-4996-son qarorida kiberxavfsizlik sohasidagi davlat siyosatini takomillashtirish chora-tadbirlari belgilab berilgan. Mazkur hujjatda kiberhujumlarni aniqlovchi va ularga javob beruvchi milliy tizimlarni yaratish, shuningdek, CERT (Computer Emergency Response Team) faoliyatini rivojlantirish orqali telekommunikatsiya tarmoqlarida yuzaga keladigan xavflarni monitoring qilish va tezkor javob mexanizmlarini yo‘lga qo‘yish vazifalari ilgari surilgan[1]. Shuningdek, O‘zbekiston Respublikasining Jinoyat kodeksi va Ma’muriy javobgarlik to‘g‘risidagi kodekslarida axborot tizimlariga noqonuniy kirish, ularni ishdan chiqarish yoki ma’lumotlarni buzish holatlari uchun javobgarlik choralarini belgilovchi moddalar mavjud (JK 2781, 278-2). Ushbu normativlar kiberhujumlarni faqat aniqlash emas, balki ularning oldini olish, huquqiy chora ko‘rish va xavfsizlikni kuchaytirish uchun ham zaruriy asos bo‘lib xizmat qiladi. Bu holatda telekommunikatsiya tarmoqlarining daxlsizligi davlat tomonidan himoya qilinadigan muhim infratuzilma sifatida e’tirof etiladi[2]. Bundan tashqari, davlat raqamli xavfsizlik siyosatida ham telekommunikatsiya tarmoqlari, mobil operatorlar, internet provayderlar va ma’lumotlar markazlari uchun xavfsizlik standartlari va texnik reglamentlar ishlab chiqilgan. Ushbu strategik hujjatlar doirasida zaifliklarni aniqlash, kiberhujumlarni monitoring qilish, kriptografik himoya vositalaridan foydalanish va axborot xavfsizligi bo‘yicha texnik chora-tadbirlarni joriy etish zaruriyati ta’kidlangan. Bu esa telekommunikatsiya tarmoqlarini himoya qilish sohasida innovatsion va tizimli yondashuvlarni tatbiq etish orqali kiberxavflarning oldini olish imkonini beradi[3]. Tadqiqot Snort va Suricata kabi tarmoq hujumlarini aniqlash tizimlarining yuqori tezlikdagi tarmoqlarda samaradorligini baholash orqali, kiber hujumlarga qarshi kurashda samarali yechimlarni topish maqsad qilingan. Ikkala tizimning eksperimental sinovlari va ularning samaradorligini tahlil qilish, zamonaviy kiber tahdidlarga javob berish imkoniyatlarini ko‘rsatadi. Ushbu tadqiqot natijalari

Management and Future Technologies journal.umft.uz June 2025 Volume 2, Issue 2 nafaqat ilmiy adabiyotga, balki amaliyotga ham katta hissa qo‘shadi, chunki u tashkilotlar va tashkilotlar uchun samarali xavfsizlik strategiyalarini ishlab chiqishga yordam beradi. Bundan tashqari, kiber xavfsizlikni ta’minlash sohasidagi o‘zgaruvchan muhit, yangi usullar va texnologiyalarning paydo bo‘lishi, tadqiqotning dolzarbligini oshiradi. Shu nuqtai nazardan, ushbu dissertatsiya zamonaviy kiber tahdidlarni bartaraf etish va tarmoq xavfsizligini oshirish bo‘yicha muhim tadqiqot hisoblanadi. Telekommunikatsiya tarmoqlariga bo‘ladigan kiber hujumlarning turlari ularning xususyatlari Telekommunikatsiya tarmoqlariga bo‘ladigan kiber hujumlar zamonaviy axborot- kommunikatsiya texnologiyalari rivojlangan bir paytda tobora ko‘payib bormoqda. Bu hujumlar turli shakllarda namoyon bo‘lib, tarmoq xavfsizligiga jiddiy tahdid soladi. Ishda eng keng tarqalgan tarmoq hujumlariga, ayniqsa, ushbu hujumlar asosida olib borilgan tajribalar davomida uchraydigan topologiyalarga alohida e’tibor qaratadi. Shuningdek, bunday hujumlarni amalga oshirish uchun ishlatiladigan asosiy vositalar haqida ma’lumot beradi. Kiberhujum - kibermakonda apparat, apparat-dasturiy va dasturiy vositalardan foydalangan holda qasddan amalga oshiriladigan, kiberxavfsizlikka tahdid soladigan harakat. Kibermakon - axborot texnologiyalari yordamida yaratilgan virtual muhit. Kiberhujumlar telekommunikatsiya tarmoqlariga turli xil usullar bilan amalga oshiriladi. Quyida ularning asosiy turlari, xususiyatlari va qo‘llanish maqsadlari keltirilgan: DDoS hujumlar - bu kompyuter tizimlariga (tarmoq resurslari yoki aloqa kanallari) ularni qonuniy foydalanuvchilar uchun imkonsiz qilishga qaratilgan hujumlar. DDoS hujumlari bir vaqtning o‘zida ma’lum bir manbaga Internetda joylashgan bir yoki bir nechta kompyuterlardan ko‘p sonli so‘rovlarni yuborishni o‘z ichiga oladi. Agar minglab, o‘n minglab millionlab kompyuterlar bir vaqtning o‘zida ma’lum bir serverga (yoki tarmoq xizmatiga) so‘rov yuborishni boshlasa, u holda server uni boshqara olmaydi yoki ushbu serverga aloqa kanali uchun tarmoqli kengligi etarli bo‘lmaydi. Ikkala holatda ham Internet foydalanuvchilari hujum qilingan serverga, hatto bloklangan aloqa kanali orqali ulangan barcha serverlar va boshqa resurslarga kira olmaydi[4]. An’anaviy DoS hujumi birinchi marta 1996-yilda, Panix internet xizmat provayderiga qilingan hujum bilan tarixda qayd etilgan. Hujum natijasida Panixning internet xizmatlari ko‘p foydalanuvchilar uchun vaqtinchalik to‘xtab qoldi. SYN flooding hujumi serverga soxta ulanish so‘rovlari yuborish orqali amalga oshirildi. Ushbu so‘rovlar serverning barcha mavjud ulanish resurslarini to‘ldirdi, natijada haqiqiy foydalanuvchilar ulana olmadilar. 279

Volume 2, Issue 2 June 2025 Management and Future Technologies journal.umft.uz 280 1-rasm. DDos hujumi amalga oshirish sxemasi Dunyo miqyosida har kuni ikki mingga yaqin DDoS hujumlari qayd etiladi. O‘rtacha DDoS hujumi yirik kompaniyaga soatiga 250 dollar zarar yetkazadi. Botnet - bu zararli dasturlar bilan infeksiya qilingan va hujumchilar tomonidan boshqariladigan ko‘p sonli kompyuterlar tarmog‘idir. Botnetlar bir vaqtning o‘zida ko‘p sonli hujumlarni amalga oshirish imkonini beradi. DDoS hujumlari odatda ko‘plab turli manbalardan keladi, bu esa hujumni aniqlash va to‘sishni qiyinlashtiradi. Bir necha ming yoki millionlab infeksiya qilingan qurilmalar bir vaqtda bir maqsadga trafik yuborishi mumkin[5]. Hujum turli protokollarda bo‘lishi mumkin: DDoS hujumlari TCP, UDP, HTTP va boshqa tarmoq protokollarida amalga oshirilishi mumkin. DDoS hujumining asosiy maqsadi maqsadli xizmatlarni to‘xtatish yoki sekinlashtirish. Bu, masalan, veb-saytlar, onlayn xizmatlar, moliyaviy operatsiyalar yoki boshqa muhim tarmoq resurslariga taalluqli bo‘ladi. Bundan tashqari, o‘rta DDoS hujumi davomiyligi odatda 6 dan 24 soatgacha davom etadi, bu da demak, har bir soat uchun $40,000 dan $100,000 gacha zarar keltirib chiqarishi mumkin. DDoS hujumlariga qarshi samarali himoya qilish uchun tarmoq infratuzilmasini monitoring qilish, trafigi filtratsiyasi va yuklanishni balanslash texnologiyalaridan foydalanish muhimdir. Shu bilan birga, tarmoqni tezda qayta tiklash va hujumlarni minimallashtirish uchun tayyorlanish ham zarur.